Robust Finite-Control-Set Model Predictive Torque Control of PMSM Drives Based on Dual Variable Bandwidth ESOs

In many aviation actuator systems, permanent magnet synchronous motor (PMSM) drives are used because of their high efficiency and compact design. Considering the aerodynamic loads in this specific application, high dynamic response performance and robustness are both essential to ensure the high maneuverability and reliability of actuator system. However, these performance requirements pose challenges for traditional approaches. In this case, a robust finite-control-set model predictive torque control (robust FCS-MPTC) based on dual variable bandwidth extended state observers (DVB-ESOs) is proposed. The dual ESOs, which are based on ultra-local discrete predictive models capable of simultaneously predicting state variables and system disturbances in a control cycle, are designed to suppress the effects of parameter mismatch. The effect of ESOs bandwidth on dynamic performance is studied in the discrete domain. Based on this, a modified cost function with a variable bandwidth technique is developed to identify the ideal voltage vector (VV) with the high dynamic performance. In addition, a weighting factor optimization strategy is presented to improve robustness. The stability of the proposed robust FCS-MPTC is analyzed. Finally, the effectiveness of the proposed method is verified by simulation and experimental results, demonstrating its superior performance when compared to several approaches.